{"id":8889,"date":"2026-11-05T13:47:15","date_gmt":"2026-11-05T06:47:15","guid":{"rendered":"https:\/\/adtek.agency\/glossary\/llms-txt\/"},"modified":"2026-11-05T13:47:15","modified_gmt":"2026-11-05T06:47:15","slug":"llms-txt","status":"publish","type":"mona_glossary","link":"https:\/\/adtek.agency\/en\/glossary\/llms-txt\/","title":{"rendered":"llms.txt"},"content":{"rendered":"

llms.txt is a Markdown file served at \/llms.txt that summarizes what a website is and links to its most important pages, so large language models (LLMs) can read it quickly. It is a community proposal published by Jeremy Howard on September 3, 2024 at llmstxt.org, not an official standard.<\/p>\n

What is llms.txt and where did it come from?<\/h2>\n

The stated purpose is to give LLM-based agents short, curated guidance plus links to detailed content. The file usually sits at the site root. The proposal also suggests offering a clean Markdown version of each important page at the same URL with .md appended (llmstxt.org, The \/llms.txt file, accessed November 2026<\/a>).<\/p>\n

llms.txt is not an XML sitemap<\/a>: a sitemap lists every URL for search engines, while llms.txt picks a few pages and describes each one in plain words.<\/p>\n

What goes into an llms.txt file<\/h2>\n

Per llmstxt.org, the file has these parts, in order:<\/p>\n

    \n
  1. An H1 (#<\/code>) with the name of the project or site. This is the only required section.<\/li>\n
  2. A blockquote (><\/code>) with a short summary.<\/li>\n
  3. Optional paragraphs of extra detail, with no headings.<\/li>\n
  4. H2 sections (##<\/code>), each a list of links in the form [name](url)<\/code>, optionally followed by a colon and notes.<\/li>\n
  5. A section named “Optional” for secondary links that an agent can skip.<\/li>\n<\/ol>\n

    Do AI platforms read llms.txt?<\/h2>\n

    In the official documentation I reviewed (November 2026), no major AI vendor confirms using llms.txt for ranking or for choosing which sources to cite.<\/p>\n